The problem that a hardware major’s sales department faced was classic, how to improve the productivity of its Key Account Management team. The sales team was 20 strong and handled about 9000 accounts. Analytics was brought in with the following objectives – Increase the conversions to sales per visit, increase ticket value and increase the number of products that the client is using. The sales were of different product groups such as Desktops, Laptops and Accessories.
The approach used was to define the probability of the next best product that each of the clients will buy. The questions that the modelling answered were – When to sell, what and to whom? The statistical approach used is called the Cox’s Proportional Hazards Model which will give the probability of a customer a buying product X at time T. The model was specified using variables that included past behaviour, and other relevant factors such as type of industry, seasonality, company size etc.
The probabilities thus derived were collated and each of the Salesmen got a list of customers to contact in the coming month with a specific product/s to offer. The list was prepared keeping the most profitable customers in focus so that the overall productivity of the team was enhanced.
The results were that over 90% of the sales calls were successful and of this about 70% picked the product that was specified for that specific client. This led to an overall improvement of sales of close to 20% and as the costs were the same as the earlier period, the additional margin was added to the bottom line.